Physics 6, 132 (2013)

Viewpoint Sharing Entanglement without Sending It

Christine Silberhorn University of Paderborn, D-33098 Paderborn, Germany Published December 4, 2013

Three new experiments demonstrate how entanglement can be shared between distant parties without the need of an entangled carrier. Subject Areas: Quantum Information

A Viewpoint on: Experimental Distribution of Entanglement with Separable Carriers A. Fedrizzi, M. Zuppardo, G. G. Gillett, M. A. Broome, M. P. Almeida, M. Paternostro, A. G. White, and T. Paterek Phys. Rev. Lett. 111, 2013 – Published December 4, 2013

Distributing Entanglement with Separable States Christian Peuntinger, Vanessa Chille, Ladislav Mišta, Jr., Natalia Korolkova, Michael Förtsch, Jan Korger, Christoph Marquardt, and Gerd Leuchs Phys. Rev. Lett. 111, 2013 – Published December 4, 2013

Experimental Entanglement Distribution by Separable States Christina E. Vollmer, Daniela Schulze, Tobias Eberle, Vitus Händchen, Jaromír Fiurášek, and Roman Schnabel Phys. Rev. Lett. 111, 2013 – Published December 4, 2013

To challenge the limited understanding of the then- tasks as quantum teleportation, efficient quantum com- young quantum theory, Einstein, Podolsky, and Rosen munication, fundamental tests of quantum physics, and constructed, in 1935, their EPR Gedankenexperiment, long-distance quantum cryptography. in which they introduced entangled states that exhibit The quantum correlations associated with entangle- strange correlations over macroscopic distances [1]. By ment are not the only type of correlations that can occur now we have learned that entangled states are an ele- in quantum systems. In order to give a rigorous def- ment of physical reality. They lie at the heart of quan- inition of the “entangled” correlations, separable states tum physics and can, in fact, be used as a powerful re- have been introduced as the converse of entanglement [5]. source in emerging quantum technologies. Yet we find Separable states are quantum states that can be prepared out in amazement that we have still not completely cap- nonlocally, and yet they can be correlated to a certain tured the full scope of the fascinating nature of entan- degree. For example, Alice might polarize one photon glement. Three different international groups have now in the vertical direction and tell Bob over the phone to reported in Physical Review Letters[2–4] experiments of polarize another photon in the horizontal direction. It is distributing entanglement between two distant parties by well confirmed that these photons (or other distant sepa- sending a nonentangled carrier. These arrangements in- rable states) cannot be subsequently entangled by means stead place the carrier in a “cheaper,” so-called separable of local operations and classical communication (LOCC) state, which exhibit correlations that can be established [6]. In this way, it seems clear that we can distinguish be- remotely between separated parties. tween entangled states as being quantum, and separable Entanglement is typically characterized by anoma- states as representing the classical world. lously strong correlations between presently noninteract- However, recent research has discovered that quantum ing parties, typically called Alice and Bob, which have physics does not stop surprising us with its counterin- normally interacted in the past. A common setup uses a tuitive properties. In 2003, Cubitt et al.[7] showed that nonlinear crystal to create an entangled pair of orthogo- sometimes separable states can actually be used to dis- nally polarized photons that are then sent separately, one tribute entanglement between Alice and Bob. This is in to Alice and the other to Bob. In the field of quantum strong contrast to any standard communication, where information science, the remote establishment of entan- the information gained by the receiver can’t be greater glement is key for most applications because it introduces than the information in the carrier. Bob, for example, purely nonclassical correlations and the counterintuitive can’t recover a color photo if all Alice sent was a black nature of quantum physics. It enables such remarkable and white copy. But quantum information is unique in

DOI: 10.1103/Physics.6.132 c 2013 American Physical Society URL: http://link.aps.org/doi/10.1103/Physics.6.132 Physics 6, 132 (2013)

that entanglement can be sent using a cheaper—more generic—form of nonclassical correlations that can be characterized by a measure called quantum discord. A perfectly classical system has zero discord, whereas quan- tum systems can have varying amounts of discord, de- pending on the mixture of possible states they occupy. Recent work has shown that quantum discord can be an important form of currency in quantum information shar- ing when three or more systems are involved [8, 9]. This raises the obvious question: Can we really observe such an entanglement distribution in experiments? Following the general idea of Cubitt et al.’s original pa- per, Alessandro Fedrizzi from the University of Queens- land, Australia, and co-workers have realized an experi- ment demonstrating the proposed entanglement distribu- tion with separable carriers using polarization-encoded FIG. 1: Three different experiments demonstrate the distribu- tion of entanglement with a nonentangled carrier. The general single photon states [2]. Their experiment illustrates scheme is to first initiate three photons or three light beams quite nicely the crux of this strange type of entangle- (A,B,C) in separable states. Alice takes A and C and puts ment distribution. The authors actually start with an them through a kind of interference process, which creates entangled pair of photons, A and B, which are shared a correlation between them, but not entanglement. The re- between Alice and Bob. However, they subsequently de- sulting C is then sent to Bob, who combines it with his B. stroy this entanglement by randomly mixing the four dif- In the final analysis, A and B are entangled, even though ferent types of possible entangled states, the so-called the carrier (C) was never entangled with either. (APS/Alan Bell states. This procedure effectively prepares a sepa- Stonebraker) rable mixed state between photons A and B, which nev- ertheless remains in a very specific form, carrying quite distinct correlations. The information carrier, photon C, many photons using the Heisenberg uncertainty relation is prepared in a similar way, yielding, again, a specific between the amplitude and phase of optical beams. This mixed state. enables a simpler preparation of more flexible three-party Once all the photons are prepared, photons A and C states as well as an easy recombination of the final state pass through a quantum gate, which is implemented by into a true two-party (bipartite) system. Following up on some sort of quantum interference between them (see Fig. this proposal, two groups—Christian Peuntinger of the 1). Altogether this procedure produces a very particular Max Planck Institute for the Science of Light, Germany, three-party state, which is specifically tailored for the and co-workers [3] and Christina Vollmer from the Albert needs of the experiment. It has the remarkable property Einstein Institute, Germany, and collaborators [4]—have that we do not find any entanglement if we analyze the implemented experiments, which establish entanglement correlations between photon C and a subsystem made by a separable state between two final light beams lo- of photons A and B. The same is true if we analyze cated at Alice’s and Bob’s stations. photon B with respect to photons A and C. Yet we do In the first case, Peuntinger et al. start with three get measurable entanglement between photon A and the independent beams, where A and C are squeezed states subsystem of photons B and C. Thus sending photon C with reduced uncertainty in amplitude or phase, and B from Alice to Bob communicates entanglement, though is the vacuum field. For the preparation of a correlated it is never entangled itself with the others. state between the beams, all three beams are manipu- Apart from gaining new insights into the fundamentals lated with small displacements in amplitude or phase, of quantum physics, one might ask about the benefits or in both. These displacements are coordinated using of the presented protocol over the direct transmission of a classical communication channel (i.e., LOCC). Once the entangled photon pairs. As stated by the authors, prepared, Alice interferes beams A and C using a beam there is no advantage if Alice and Bob have a maximally splitter, and she then sends the resulting C beam to Bob, entangled state at their disposal. However, if the photon who interferes it with his beam B in another beam split- source is noisy, such entangled states are hard to produce, ter (see Fig. 1). Throughout the protocol, beam C re- whereas Fedrizzi et al. show that the information carried mains separable from the other beams, but at the end, by a separable photon would be more resilient against beams A and B share true bipartite entanglement. noise in some cases. In this way, Peuntinger et al. have realized a proto- Although simple to imagine, single photons acting as col with the special feature that it combines a quantum binary qubits are difficult to prepare and to work with. channel with classical communication between Alice and A possible simplification is to translate the original pro- Bob. Surprisingly, Bob can even get entanglement a pos- tocol into the framework of Gaussian states, as recently teriori, by performing the classically communicated ma- proposed in [10]. Here the information is encoded on nipulations on beam B after it has interfered with beam

DOI: 10.1103/Physics.6.132 c 2013 American Physical Society URL: http://link.aps.org/doi/10.1103/Physics.6.132 Physics 6, 132 (2013)

C in the beam splitter. This highlights interplay between References classical and quantum correlations and the role of sepa- rable states as quantum carriers. [1] A. Einstein, B. Podolsky, and N. Rosen, “Can Quantum- Vollmer and co-workers use a very similar system with Mechanical Description of Physical Reality Be Considered the same operating principle as the one just described. Complete?” Phys. Rev. 47, 777 (1935). [2] A. Fedrizzi, M. Zuppardo, G. G. Gillett, M. A. Broome, M. The main difference is that the initial state is not di- P. Almeida, M. Paternostro, A. G. White, and T. Paterek, rectly prepared by LOCC. The team starts with an en- “Experimental Distribution of Entanglement with Separable tangled state and “hides” the entanglement by mixing Carriers,” Phys. Rev. Lett. 111, 230504 (2013). one part of the system with a noisy thermal state. In [3] C. Peuntinger, V. Chille, L. Mišta, N. Korolkova, M. Förtsch, their analysis, they investigate which features are needed J. Korger, C. Marquardt, and G. Leuchs, “Distributing Entan- glement with Separable States,” Phys. Rev. Lett. 111, 230506 to see the wanted effects. This reveals that only a spe- (2013). cific class of entangled Gaussian states can be used for [4] C. E. Vollmer, D. Schulze, T. Eberle, V. Händchen, J. Fi- the preparation of the strange three-party state. Here, urášek, and R. Schnabel, “Experimental Entanglement Dis- it is important that one starts with not very clean—this tribution by Separable States,” Phys. Rev. Lett. 111, 230505 means mixed—quantum states, but the actual squeezing (2013). [5] R. F. Werner, “Quantum States with Einstein-Podolsky- of the initial states, which is the more apparent quantum Rosen Correlations Admitting a Hidden-Variable Model,” property, is less relevant. Phys. Rev. A 40, 4277 (1989). In summary, all three experiments highlight that quan- [6] C. H. Bennett, D. P. DiVincenzo, J. A. Smolin, and W. K. tum communication really becomes more subtle if we use Wootters, “Mixed-state Entanglement And Quantum Error more than two photons or beams to create multimode Correction,” Phys. Rev. A 54, 3824 (1996). [7] T. S. Cubitt, F. Verstraete, W. Dür, and J. I. Cirac, “Separa- states. This is particularly interesting if we extend quan- ble States Can Be Used To Distribute Entanglement,” Phys. tum communication beyond two parties and into more Rev. Lett. 91, 037902 (2003). complex quantum network structures. The experiments [8] A. Steltsov, H. Kampermann, and D. Bruß, “Quantum Cost also provide a test bed for studying the interplay of en- for Sending Entanglement,” Phys. Rev. Lett. 108, 250501 (2012). tanglement, noise, and classical communication. Future [9] T. K. Chuan, J. Maillard, K. Modi, T. Paterek, M. Paternos- work in this area should give a deeper understanding of tro, and M. Piani, “Quantum Discord Bounds the Amount the advantages of using separable states for entanglement of Distributed Entanglement,” Phys. Rev. Lett. 109, 070501 distribution between Alice and Bob, as well as in more (2012). complex network structures. [10] L. Mišta, Jr. and N. Korolkova, “Distribution of Continuous- Variable Entanglement by Separable Gaussian States,” Phys. Rev A 77, 050302 (2008).

About the Author Christine Silberhorn

Christine Silberhorn is a professor at the University of Paderborn, where she is leading a research group in the area of integrated quantum optics. Her interests cover novel optical technologies based on quantum optics, and light-based quantum systems for use in quantum communication and quantum information processing. She has contributed to the develop- ment of engineered quantum light sources using integrated optics and ultrafast pulsed lasers, the implementation of multichannel quantum networks for photon counting and quantum simulations, and the realization of quantum communication systems with bright light. She received her doctorate from the University of Erlangen in 2003, and worked as a postdoc at the University of Oxford from 2003 to 2004. From 2005 to 2010 she was a Max Planck Research Group Leader in Erlangen. (Photo by DFG Ausserhofer.)

DOI: 10.1103/Physics.6.132 c 2013 American Physical Society URL: http://link.aps.org/doi/10.1103/Physics.6.132